I don't think that Voinovich wins Minnesota (the state swings little), so it is really close. He is more likely to win Michigan than Minnesota. Voinovich does not need help in Florida. He has more legitimate achievements than Dubya, and it shows.
I can see him winning Pennsylvania.
IOTL, Bush came closer to winning Minnesota than Michigan.
In the worst two years for Democratic nominees for President, Minnesota was second-best and best in those two elections (1972 and 1984). In 2008, the best year for a Democratic nominee for President in a binary election since LBJ in 1964, it wasn't anything special.
But that is one state, one that swings less than any other state. It's also nearly trivial.
In a Democratic blowout (60-40) the Democrat could win the state 59-41.
...We have two different meanings for the word "swing" in the Atlas. A "swing state" is a state that is typically close in a close election and can make the difference in a close election -- like Colorado, Florida, Iowa, New Hampshire, or Ohio. A state that can swing wildly is one that goes very badly for one nominee, exaggerating his loss and going very well for someone who fits the culture. Example: West Virginia was worse than average for McGovern, but really good for Carter and Clinton, and absolutely horrible for Obama.
I thought that Voinovich was going to do better in the general election and win with less controversy. A scenario in which Voinovich wins without Florida is intriguing.